{"data":{"kind":"file","path":"README.md","version_id":"t5kbr1akkps9iokcbub5l5qb","entry":{"name":"README.md","path":"README.md","is_directory":false,"size":1956,"modified_at":"2026-06-10T06:40:33.980000","content_hash":"6728abf9b7ca74a9c88276a8e7dccc57bcecb7c4f1f43cd66a40903baaff6b2d"},"entries":[],"content":"# tic-tac-toe\n\n### Overview\n- **Environment ID**: `tic-tac-toe`\n- **Short description**: Multi-turn Tic-Tac-Toe tool environment with configurable text or image board state.\n- **Tags**: `multi-turn`, `tools`, `game`, `train`, `eval`\n\n### Datasets\n- **Primary dataset(s)**: Synthetic repeated starting-board examples generated in `load_environment`.\n- **Source links**: None.\n- **Split sizes**: Controlled by `num_examples` and defaults to 100 examples.\n\n### Task\n- **Type**: Multi-turn tool use.\n- **Output format expectations**: The model should call the exposed move tool with a board position from 0 to 8.\n- **Rubric overview**: `win_reward` returns 1.0 when X wins and 0.0 otherwise.\n\n### Quickstart\nRun an image-board evaluation with default settings:\n\n```bash\nprime eval run tic-tac-toe\n```\n\nRun a text-only evaluation:\n\n```bash\nprime eval run tic-tac-toe -a '{\"mode\": \"text\"}'\n```\n\nRun an image-only evaluation explicitly:\n\n```bash\nprime eval run tic-tac-toe -a '{\"mode\": \"image\"}'\n```\n\nConfigure model and sampling:\n\n```bash\nprime eval run tic-tac-toe -m gpt-4.1-mini -n 20 -r 3 -t 1024 -T 0.7 -a '{\"mode\": \"text\"}'\n```\n\n### Environment Arguments\nUse `-a` / `--env-args` to pass environment-specific configuration as a JSON object.\n\n| Arg | Type | Default | Description |\n| --- | ---- | ------- | ----------- |\n| `mode` | str | `\"image\"` | Board modality. Use `\"text\"` for ASCII boards, `\"image\"` for PNG board images, or `\"mixed\"` for alternating rows. Aliases include `\"visual\"` for image mode and `\"ascii\"` for text mode. |\n| `num_examples` | int | `100` | Number of repeated starting-board examples to include. Must be at least 1. |\n| `mixed_batch` | bool | `null` | Deprecated compatibility flag. `true` maps to `mode=\"mixed\"`; `false` maps to `mode=\"image\"`. |\n\n### Metrics\n\n| Metric | Meaning |\n| ------ | ------- |\n| `reward` | Main scalar reward; 1.0 when X wins, 0.0 otherwise. |\n| `win_reward` | Raw binary win signal from the rubric. |\n","encoding":"utf-8","truncated":false,"total_bytes":1956},"status":null}